Skip to content

P
Postgres FD Implementation
Commit f7a839bc authored by Tom Lane's avatar Tom Lane
Browse files
Options

Clean up portability problems in regexp package: change all routine 

definitions from K&R to ANSI C style, and fix broken assumption that
int and long are the same datatype.  This repairs problems observed
on Alpha with regexps having between 32 and 63 states.
parent f4e4c729
Expand all Hide whitespace changes
Inline Side-by-side
Showing with 226 additions and 603 deletions
src/backend/regex/engine.c
View file @ f7a839bc
......@@ -80,37 +80,26 @@ struct match
pg_wchar *endp; /* end of string -- virtual NUL here */
pg_wchar *coldp; /* can be no match starting before here */
pg_wchar **lastpos; /* [nplus+1] */
STATEVARS;
STATEVARS;
states st; /* current states */
states fresh; /* states for a fresh start */
states tmp; /* temporary */
states empty; /* empty set of states */
};
/* ========= begin header generated by ./mkh ========= */
#ifdef __cplusplus
extern "C"
{
#endif
static int matcher(struct re_guts *g, pg_wchar *string, size_t nmatch,
regmatch_t *pmatch, int eflags);
static pg_wchar *dissect(struct match *m, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst);
static pg_wchar *backref(struct match *m, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst, sopno lev);
static pg_wchar *fast(struct match *m, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst);
static pg_wchar *slow(struct match *m, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst);
static states step(struct re_guts *g, sopno start,
sopno stop, states bef, int ch, states aft);
/* === engine.c === */
static int
matcher(struct re_guts * g, pg_wchar * string, size_t nmatch,
regmatch_t *pmatch, int eflags);
static pg_wchar *
dissect(struct match * m, pg_wchar * start, pg_wchar * stop,
sopno startst, sopno stopst);
static pg_wchar *
backref(struct match * m, pg_wchar * start, pg_wchar * stop,
sopno startst, sopno stopst, sopno lev);
static pg_wchar *
fast(struct match * m, pg_wchar * start, pg_wchar * stop,
sopno startst, sopno stopst);
static pg_wchar *
slow(struct match * m, pg_wchar * start, pg_wchar * stop, sopno startst, sopno stopst);
static states
step(struct re_guts * g, sopno start,
sopno stop, states bef, int ch, states aft);
#define BOL (OUT+1)
#define EOL (BOL+1)
#define BOLEOL (BOL+2)
......@@ -128,24 +117,13 @@ extern "C"
#endif
#ifdef REDEBUG
static void
print(struct match * m, pg_wchar * caption, states st, int ch, FILE *d);
#endif
#ifdef REDEBUG
static void
at(struct match * m, pg_wchar * title, pg_wchar * start, pg_wchar * stop,
sopno startst, sopno stopst);
#endif
#ifdef REDEBUG
static pg_wchar *
p_char(int ch);
#endif
#ifdef __cplusplus
}
static void print(struct match *m, pg_wchar *caption, states st, int ch,
FILE *d);
static void at(struct match *m, pg_wchar *title, pg_wchar *start,
pg_wchar *stop, sopno startst, sopno stopst);
static pg_wchar *pchar(int ch);
static int pg_isprint(int c);
#endif
/* ========= end header generated by ./mkh ========= */
#ifdef REDEBUG
#define SP(t, s, c) print(m, t, s, c, stdout)
......@@ -158,17 +136,11 @@ extern "C"
#endif
/*
- matcher - the actual matching engine
== static int matcher(struct re_guts *g, pg_wchar *string, \
== size_t nmatch, regmatch_t *pmatch, int eflags);
* matcher - the actual matching engine
*/
static int /* 0 success, REG_NOMATCH failure */
matcher(g, string, nmatch, pmatch, eflags)
struct re_guts *g;
pg_wchar *string;
size_t nmatch;
regmatch_t *pmatch;
int eflags;
matcher(struct re_guts *g, pg_wchar *string, size_t nmatch,
regmatch_t *pmatch, int eflags)
{
pg_wchar *endp;
int i;
......@@ -206,10 +178,11 @@ int eflags;
for (dp = start; dp < stop; dp++)
if (*dp == g->must[0] && stop - dp >= g->mlen &&
#ifdef MULTIBYTE
memcmp(dp, g->must, (size_t) (g->mlen * sizeof(pg_wchar))) == 0)
memcmp(dp, g->must, (size_t) (g->mlen * sizeof(pg_wchar))) == 0
#else
memcmp(dp, g->must, (size_t) g->mlen) == 0)
memcmp(dp, g->must, (size_t) g->mlen) == 0
#endif
)
break;
if (dp == stop) /* we didn't find g->must */
return REG_NOMATCH;
......@@ -349,17 +322,11 @@ int eflags;
}
/*
- dissect - figure out what matched what, no back references
== static char *dissect(struct match *m, char *start, \
== char *stop, sopno startst, sopno stopst);
* dissect - figure out what matched what, no back references
*/
static pg_wchar * /* == stop (success) always */
dissect(m, start, stop, startst, stopst)
struct match *m;
pg_wchar *start;
pg_wchar *stop;
sopno startst;
sopno stopst;
dissect(struct match *m, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst)
{
int i;
sopno ss; /* start sop of current subRE */
......@@ -549,18 +516,13 @@ sopno stopst;
}
/*
- backref - figure out what matched what, figuring in back references
== static char *backref(struct match *m, char *start, \
== char *stop, sopno startst, sopno stopst, sopno lev);
* backref - figure out what matched what, figuring in back references
*
* lev is PLUS nesting level
*/
static pg_wchar * /* == stop (success) or NULL (failure) */
backref(m, start, stop, startst, stopst, lev)
struct match *m;
pg_wchar *start;
pg_wchar *stop;
sopno startst;
sopno stopst;
sopno lev; /* PLUS nesting level */
backref(struct match *m, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst, sopno lev)
{
int i;
sopno ss; /* start sop of current subRE */
......@@ -763,17 +725,11 @@ sopno lev; /* PLUS nesting level */
}
/*
- fast - step through the string at top speed
== static char *fast(struct match *m, char *start, \
== char *stop, sopno startst, sopno stopst);
* fast - step through the string at top speed
*/
static pg_wchar * /* where tentative match ended, or NULL */
fast(m, start, stop, startst, stopst)
struct match *m;
pg_wchar *start;
pg_wchar *stop;
sopno startst;
sopno stopst;
fast(struct match *m, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst)
{
states st = m->st;
states fresh = m->fresh;
......@@ -858,17 +814,11 @@ sopno stopst;
}
/*
- slow - step through the string more deliberately
== static char *slow(struct match *m, char *start, \
== char *stop, sopno startst, sopno stopst);
* slow - step through the string more deliberately
*/
static pg_wchar * /* where it ended */
slow(m, start, stop, startst, stopst)
struct match *m;
pg_wchar *start;
pg_wchar *stop;
sopno startst;
sopno stopst;
slow(struct match *m, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst)
{
states st = m->st;
states empty = m->empty;
......@@ -948,27 +898,15 @@ sopno stopst;
/*
- step - map set of states reachable before char to set reachable after
== static states step(struct re_guts *g, sopno start, sopno stop, \
== states bef, int ch, states aft);
== #define BOL (OUT+1)
== #define EOL (BOL+1)
== #define BOLEOL (BOL+2)
== #define NOTHING (BOL+3)
== #define BOW (BOL+4)
== #define EOW (BOL+5)
== #define CODEMAX (BOL+5) // highest code used
== #define NONCHAR(c) ((c) > CHAR_MAX)
== #define NNONCHAR (CODEMAX-CHAR_MAX)
* step - map set of states reachable before char to set reachable after
*/
static states
step(g, start, stop, bef, ch, aft)
struct re_guts *g;
sopno start; /* start state within strip */
sopno stop; /* state after stop state within strip */
states bef; /* states reachable before */
int ch; /* character or NONCHAR code */
states aft; /* states already known reachable after */
step(struct re_guts *g,
sopno start, /* start state within strip */
sopno stop, /* state after stop state within strip */
states bef, /* states reachable before */
int ch, /* character or NONCHAR code */
states aft) /* states already known reachable after */
{
cset *cs;
sop s;
......@@ -1082,19 +1020,11 @@ states aft; /* states already known reachable after */
#ifdef REDEBUG
/*
- print - print a set of states
== #ifdef REDEBUG
== static void print(struct match *m, char *caption, states st, \
== int ch, FILE *d);
== #endif
* print - print a set of states
*/
static void
print(m, caption, st, ch, d)
struct match *m;
pg_wchar *caption;
states st;
int ch;
FILE *d;
print(struct match *m, pg_wchar *caption, states st,
int ch, FILE *d)
{
struct re_guts *g = m->g;
int i;
......@@ -1116,20 +1046,11 @@ FILE *d;
}
/*
- at - print current situation
== #ifdef REDEBUG
== static void at(struct match *m, pg_wchar *title, pg_wchar *start, pg_wchar *stop, \
== sopno startst, sopno stopst);
== #endif
* at - print current situation
*/
static void
at(m, title, start, stop, startst, stopst)
struct match *m;
pg_wchar *title;
pg_wchar *start;
pg_wchar *stop;
sopno startst;
sopno stopst;
at(struct match *m, pg_wchar *title, pg_wchar *start, pg_wchar *stop,
sopno startst, sopno stopst)
{
if (!(m->eflags & REG_TRACE))
return;
......@@ -1140,19 +1061,26 @@ sopno stopst;
}
#ifndef PCHARDONE
#define PCHARDONE /* never again */
#define PCHARDONE /* only do this once */
/*
- pchar - make a character printable
== #ifdef REDEBUG
== static char *pchar(int ch);
== #endif
* pchar - make a character printable
*
* Is this identical to regchar() over in debug.c? Well, yes. But a
* duplicate here avoids having a debugging-capable regexec.o tied to
* a matching debug.o, and this is convenient. It all disappears in
* the non-debug compilation anyway, so it doesn't matter much.
*/
static pg_wchar * /* -> representation */
pchar(int ch)
{
static pg_wchar pbuf[10];
if (pg_isprint(ch) || ch == ' ')
sprintf(pbuf, "%c", ch);
else
sprintf(pbuf, "\\%o", ch);
return pbuf;
}
static int
pg_isprint(int c)
......@@ -1164,19 +1092,6 @@ pg_isprint(int c)
#endif
}
static pg_wchar * /* -> representation */
pchar(ch)
int ch;
{
static pg_wchar pbuf[10];
if (pg_isprint(ch) || ch == ' ')
sprintf(pbuf, "%c", ch);
else
sprintf(pbuf, "\\%o", ch);
return pbuf;
}
#endif
#endif
......
src/backend/regex/regcomp.c
View file @ f7a839bc
This diff is collapsed.
src/backend/regex/regerror.c
View file @ f7a839bc
......@@ -37,59 +37,19 @@
* @(#)regerror.c 8.4 (Berkeley) 3/20/94
*/
#if defined(LIBC_SCCS) && !defined(lint)
static char sccsid[] = "@(#)regerror.c 8.4 (Berkeley) 3/20/94";
#endif /* LIBC_SCCS and not lint */
#include "postgres.h"
#include <sys/types.h>
#include <stdio.h>
#include <string.h>
#include <ctype.h>
#include <limits.h>
#include <stdlib.h>
#include <assert.h>
#include "regex/regex.h"
#include "regex/utils.h"
#include "regex/regex2.h"
/* ========= begin header generated by ./mkh ========= */
#ifdef __cplusplus
extern "C"
{
#endif
/* === regerror.c === */
static char *regatoi(const regex_t *preg, char *localbuf);
static char *regatoi(const regex_t *preg, char *localbuf);
#ifdef __cplusplus
}
#endif
/* ========= end header generated by ./mkh ========= */
/*
= #define REG_NOMATCH 1
= #define REG_BADPAT 2
= #define REG_ECOLLATE 3
= #define REG_ECTYPE 4
= #define REG_EESCAPE 5
= #define REG_ESUBREG 6
= #define REG_EBRACK 7
= #define REG_EPAREN 8
= #define REG_EBRACE 9
= #define REG_BADBR 10
= #define REG_ERANGE 11
= #define REG_ESPACE 12
= #define REG_BADRPT 13
= #define REG_EMPTY 14
= #define REG_ASSERT 15
= #define REG_INVARG 16
= #define REG_ATOI 255 // convert name to number (!)
= #define REG_ITOA 0400 // convert number to name (!)
*/
static struct rerr
{
int code;
......@@ -152,16 +112,12 @@ static struct rerr
};
/*
- regerror - the interface to error numbers
= extern size_t regerror(int, const regex_t *, char *, size_t);
* regerror - the interface to error numbers
*/
/* ARGSUSED */
size_t
pg95_regerror(errcode, preg, errbuf, errbuf_size)
int errcode;
const regex_t *preg;
char *errbuf;
size_t errbuf_size;
pg95_regerror(int errcode, const regex_t *preg,
char *errbuf, size_t errbuf_size)
{
struct rerr *r;
size_t len;
......@@ -206,13 +162,10 @@ size_t errbuf_size;
}
/*
- regatoi - internal routine to implement REG_ATOI
== static char *regatoi(const regex_t *preg, char *localbuf);
* regatoi - internal routine to implement REG_ATOI
*/
static char *
regatoi(preg, localbuf)
const regex_t *preg;
char *localbuf;
regatoi(const regex_t *preg, char *localbuf)
{
struct rerr *r;
......
src/backend/regex/regexec.c
View file @ f7a839bc
......@@ -37,11 +37,6 @@
* @(#)regexec.c 8.3 (Berkeley) 3/20/94
*/
#if defined(LIBC_SCCS) && !defined(lint)
static char sccsid[] = "@(#)regexec.c 8.3 (Berkeley) 3/20/94";
#endif /* LIBC_SCCS and not lint */
#include "postgres.h"
/*
......@@ -52,9 +47,6 @@ static char sccsid[] = "@(#)regexec.c 8.3 (Berkeley) 3/20/94";
* representations for state sets.
*/
#include <sys/types.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <limits.h>
#include <ctype.h>
#include <assert.h>
......@@ -69,24 +61,24 @@ static int nope = 0; /* for use in asserts; shuts lint up */
#define states long
#define states1 states /* for later use in regexec() decision */
#define CLEAR(v) ((v) = 0)
#define SET0(v, n) ((v) &= ~(1 << (n)))
#define SET1(v, n) ((v) |= 1 << (n))
#define ISSET(v, n) ((v) & (1 << (n)))
#define SET0(v, n) ((v) &= ~(1L << (n)))
#define SET1(v, n) ((v) |= (1L << (n)))
#define ISSET(v, n) ((v) & (1L << (n)))
#define ASSIGN(d, s) ((d) = (s))
#define EQ(a, b) ((a) == (b))
#define STATEVARS int dummy /* dummy version */
#define STATESETUP(m, n) /* nothing */
#define STATETEARDOWN(m) /* nothing */
#define SETUP(v) ((v) = 0)
#define onestate int
#define INIT(o, n) ((o) = (unsigned)1 << (n))
#define INC(o) ((o) <<= 1)
#define ISSTATEIN(v, o) ((v) & (o))
#define onestate long
#define INIT(o, n) ((o) = (1L << (n)))
#define INC(o) ((o) <<= 1)
#define ISSTATEIN(v, o) ((v) & (o))
/* some abbreviations; note that some of these know variable names! */
/* do "if I'm here, I can also be there" etc without branches */
#define FWD(dst, src, n) ((dst) |= ((unsigned)(src)&(here)) << (n))
#define BACK(dst, src, n) ((dst) |= ((unsigned)(src)&(here)) >> (n))
#define ISSETBACK(v, n) ((v) & ((unsigned)here >> (n)))
#define FWD(dst, src, n) ((dst) |= ((src) & (here)) << (n))
#define BACK(dst, src, n) ((dst) |= ((src) & (here)) >> (n))
#define ISSETBACK(v, n) ((v) & (here >> (n)))
/* function names */
#define SNAMES /* engine.c looks after details */
......@@ -129,7 +121,7 @@ static int nope = 0; /* for use in asserts; shuts lint up */
#define SETUP(v) ((v) = &m->space[m->vn++ * m->g->nstates])
#define onestate int
#define INIT(o, n) ((o) = (n))
#define INC(o) ((o)++)
#define INC(o) ((o)++)
#define ISSTATEIN(v, o) ((v)[o])
/* some abbreviations; note that some of these know variable names! */
/* do "if I'm here, I can also be there" etc without branches */
......@@ -142,27 +134,14 @@ static int nope = 0; /* for use in asserts; shuts lint up */
#include "engine.c"
/*
- regexec - interface for matching
= extern int regexec(const regex_t *, const char *, size_t, \
= regmatch_t [], int);
= #define REG_NOTBOL 00001
= #define REG_NOTEOL 00002
= #define REG_STARTEND 00004
= #define REG_TRACE 00400 // tracing of execution
= #define REG_LARGE 01000 // force large representation
= #define REG_BACKR 02000 // force use of backref code
* regexec - interface for matching
*
* We put this here so we can exploit knowledge of the state representation
* when choosing which matcher to call. Also, by this point the matchers
* have been prototyped.
* when choosing which matcher to call.
*/
int /* 0 success, REG_NOMATCH failure */
pg95_regexec(preg, string, nmatch, pmatch, eflags)
const regex_t *preg;
const char *string;
size_t nmatch;
regmatch_t *pmatch;
int eflags;
pg95_regexec(const regex_t *preg, const char *string, size_t nmatch,
regmatch_t *pmatch, int eflags)
{
struct re_guts *g = preg->re_g;
......
src/backend/regex/regfree.c
View file @ f7a839bc
......@@ -37,28 +37,19 @@
* @(#)regfree.c 8.3 (Berkeley) 3/20/94
*/
#if defined(LIBC_SCCS) && !defined(lint)
static char sccsid[] = "@(#)regfree.c 8.3 (Berkeley) 3/20/94";
#endif /* LIBC_SCCS and not lint */
#include "postgres.h"
#include <sys/types.h>
#include <stdio.h>
#include <stdlib.h>
#include "regex/regex.h"
#include "regex/utils.h"
#include "regex/regex2.h"
/*
- regfree - free everything
= extern void regfree(regex_t *);
* regfree - free everything
*/
void
pg95_regfree(preg)
regex_t *preg;
pg95_regfree(regex_t *preg)
{
struct re_guts *g;
......
src/include/regex/cclass.h
View file @ f7a839bc
......@@ -94,6 +94,6 @@ static struct cclass
""
},
{
NULL, 0, ""
NULL, NULL, ""
}
};
src/include/regex/regex.h
View file @ f7a839bc
......@@ -102,10 +102,12 @@ typedef struct
#define REG_LARGE 01000 /* force large representation */
#define REG_BACKR 02000 /* force use of backref code */
int pg95_regcomp(regex_t *, const char *, int);
size_t pg95_regerror(int, const regex_t *, char *, size_t);
int pg95_regexec(const regex_t *,
const char *, size_t, regmatch_t[], int);
void pg95_regfree(regex_t *);
extern int pg95_regcomp(regex_t *preg, const char *pattern, int cflags);
extern size_t pg95_regerror(int errcode, const regex_t *preg,
char *errbuf, size_t errbuf_size);
extern int pg95_regexec(const regex_t *preg, const char *string,
size_t nmatch,
regmatch_t *pmatch, int eflags);
extern void pg95_regfree(regex_t *preg);
#endif /* !_REGEX_H_ */
src/include/regex/regex2.h
View file @ f7a839bc
......@@ -39,22 +39,6 @@
#include <limits.h>
/*
* First, the stuff that ends up in the outside-world include file
*/
/*
typedef off_t regoff_t;
typedef struct {
int re_magic;
size_t re_nsub; // number of parenthesized subexpressions
const char *re_endp; // end pointer for REG_PEND
struct re_guts *re_g; // none of your business :-)
} regex_t;
typedef struct {
regoff_t rm_so; // start of match
regoff_t rm_eo; // end of match
} regmatch_t;
*/
/*
* internals of regex_t
*/
......@@ -82,8 +66,8 @@
typedef unsigned long sop; /* strip operator */
typedef long sopno;
#define OPRMASK 0xf8000000
#define OPDMASK 0x07ffffff
#define OPRMASK ((sop) 0xf8000000)
#define OPDMASK ((sop) 0x07ffffff)
#define OPSHIFT ((unsigned)27)
#define OP(n) ((n)&OPRMASK)
#define OPND(n) ((n)&OPDMASK)
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment